在2d阵列的Numpy卷垂直

Question

如何(有效地)执行以下操作:

x = np.arange(49)
x2 = np.reshape(x, (7,7))

x2
array([[ 0,  1,  2,  3,  4,  5,  6],
       [ 7,  8,  9, 10, 11, 12, 13],
       [14, 15, 16, 17, 18, 19, 20],
       [21, 22, 23, 24, 25, 26, 27],
       [28, 29, 30, 31, 32, 33, 34],
       [35, 36, 37, 38, 39, 40, 41],
       [42, 43, 44, 45, 46, 47, 48]])

从这里开始我想推出几件事.我想掷0,7,14,21等所以14来到顶部.然后与4,11,18,25等相同,所以39来到顶部.
结果应该是:

x2
array([[14,  1,  2,  3, 39,  5,  6],
       [21,  8,  9, 10, 46, 12, 13],
       [28, 15, 16, 17,  4, 19, 20],
       [35, 22, 23, 24, 11, 26, 27],
       [42, 29, 30, 31, 18, 33, 34],
       [ 0, 36, 37, 38, 25, 40, 41],
       [ 7, 43, 44, 45, 32, 47, 48]])

我查了numpy.roll,这里和google但是找不到怎么会这样做.
对于水平滚动,我可以这样做:

np.roll(x2[0], 3, axis=0)

x3
array([4, 5, 6, 0, 1, 2, 3])

但是如何将此卷更改作为新副本返回完整数组？

Answer 1

滚动负移：

x2[:, 0] = np.roll(x2[:, 0], -2)

滚动正向移动：

x2[:, 4] = np.roll(x2[:, 4], 2)

给出：

>>>x2
array([[14,  1,  2,  3, 39,  5,  6],
       [21,  8,  9, 10, 46, 12, 13],
       [28, 15, 16, 17,  4, 19, 20],
       [35, 22, 23, 24, 11, 26, 27],
       [42, 29, 30, 31, 18, 33, 34],
       [ 0, 36, 37, 38, 25, 40, 41],
       [ 7, 43, 44, 45, 32, 47, 48]])

Answer 2

您必须覆盖该列

例如：

x2[:,0] = np.roll(x2[:,0], 3)